Skip to main content

add_job

Create or update a training job. Learn more about training jobs.

Parameters​

mb.add_job({training_function}, python_version=, ...)

Standard parameters​

  • training_function: Callable The function used to define the job. It should return a trained model object, or similar.
  • name: Optional[str]: The name of the training job. If no name is supplied the name of training_function will be used.
  • branch: Optional[str]: The branch where the job will be stored. By default it's the current branch of the session.
  • python_version: Optional["3.6"|"3.7"|"3.8"|"3.9"|"3.10"|"3.11"|"3.12"] Python version to run in production.
  • python_packages: Optional[List[str]] The pip packages to install in the training job's environment. E.g. ["scikit-learn==1.0.2"].
  • requirements_txt_path: Optional[str] The filepath of a pip-compatible requirements.txt that specifies all of the packages needed for this training job.
  • system_packages: Optional[List[str]] The apt-get packages to install in the training job's environment. E.g. ["libgomp1"].

Extra files parameters​

  • extra_files: Optional[str, List[str], Dict[str, str]] Additional files to include with the training job.
    • as str: The path to a single file or directory.
    • as List[str]: Paths to multiple files or directories.
    • as Dict[str, str]: Pairs of file paths or directories. The key is the local file path, the value is the file path in the training job. This format is for advanced use cases needing name or path of a file to be different in the training job vs. locally.
  • skip_extra_files_dependencies: Optional[bool] If True then mb.add_job will not parse extra_files to determine additional Python package dependencies. Default is False.
  • skip_extra_files_discovery: Optional[bool] If True then mb.add_job will not include additional files it discovers while parsing the files listed extra_files. Use this to limit mb.add_job to only include the extra_files that are specified. Default is False.

Returns​

The call to mb.add_job is typically done in an interactive environment, like a Python notebook. The command will print out status, and then show link to open the job in Modelbit.

Examples​

This example function trains a model and adds it to the model registry.

def train_model():
  # code to create a model
  my_model = ...

  # add the model to the registry
  mb.add_model(my_model)

Creating a training job​

Create a training job that will run the function, train_model:

mb.add_job(train_model)

Including dependencies​

Specify the Python version, extra files, or additional packages to install when creating the training job:

mb.add_job(train_model,
           name="my_training_job",
           python_version="3.10",
           python_packages=["pandas==2.2.2"],
           extra_files=["some_file.txt"])

Deploying with a requirements.txt file​

Specify the path to a pip-compatible requirements.txt file that defines your custom Python environment with requirements_txt_path= :

mb.add_job(train_model, python_version="3.10", requirements_txt_path="my_requirements.txt")

See also​